Near-field multiple source localization by passive sensor array

Huang, Yufei; Barkat, M.

doi:10.1109/8.86917

Cited by 355 publications

(172 citation statements)

References 14 publications

Supporting

Mentioning

171

Contrasting

Unclassified

Order By: Relevance

“…[1][2][3][4][5][6][7][8][9] 초기 다양한 위치 추정기법들은 대부분 원거리(far 최근 선형 선배열에서 근거리 위치 추정에 관한 연구가 많이 진행되었다. [3][4][5][6][7][8] 그중 가장 일반적인 기 법으로는 1차원 탐색 기법을 2차원으로 확장한 ML, [3] 2차원 MUSIC [4] 기법이 있으나 연산량이 많이 필요한 단점이 있다. 2차원 탐색을 피하기 위해 선형 예측 기법(linear prediction), [5] higher-order ESPRIT, [6] cumulant 기반 [7] 등의 기법이 제안되었지만 추가적으로 …”

Section: 서 론unclassified

Near-field Source Localization Method using Matrix Pencil

Jung¹,

Lee²,

Yoon³

et al. 2013

The Journal of the Acoustical Society of Korea

View full text Add to dashboard Cite

ABSTRACT:In this paper, near-field source localization algorithm is presented using Matrix Pencil in Uniform Linear Array(ULA). Based on the centrosymmetry of the ULA, the proposed algorithm decouples the steering vectors which allow for the bearing estimation using Matrix pencil. With estimated bearing, the range estimation of each source is consequently obtained by defining 1D MUSIC spectrum. Simulation results are presented to validate the performance of the proposed algorithm. 예측 기법(linear prediction), [5] higher-order ESPRIT, [6] cumulant 기반 [7] 등의 기법이 제안되었지만 추가적으로 II. 신호 모델

show abstract

Section: 서 론unclassified

Near-field Source Localization Method using Matrix Pencil

Jung¹,

Lee²,

Yoon³

et al. 2013

The Journal of the Acoustical Society of Korea

View full text Add to dashboard Cite

show abstract

Section: Computational Complexitymentioning

confidence: 99%

“…Although these algorithms in [13,21,22,25] also have not parameters match procedure, these algorithms have large amount of computational cost for the high-dimensional search or complex iterative process. More specifically, the algorithm in [13] needs to perform 2D search process for localizing the 2D NF sources, the algorithms in [21,22] performs 3D search process for localizing the 3D NF sources, and the algorithm in [25] requires complex iterative procedure. However, the proposed algorithm needs to form two (4M −1)×(4M −1) covariance matrixes for Z and Y axes respectively, to construct two (4M −1)×(4M −1) Toeplitz matrixes for Z and Y axes respectively, to perform eigendecomposition of the Toeplitz matrix, and to execute once one-dimensional search for the elevation angle estimation.…”

Section: Computational Complexitymentioning

confidence: 99%

“…More specifically, for 2D NF sources localization, Swindlehurst and Kailath proposed a spatial Wigner distribution approach [12], however, this method has a less-than-ideal estimation accuracy. Later on, Huang and Barkat [13] developed a 2D MUSIC method with high resolution, but it requires 2D search and thus has huge amount of computations. In addition, some other algorithms, such as the higher order ESPRITLike algorithm [14], the subspace-based (SB) algorithm [15], the two steps (TS) algorithm [16], and the weighted linear prediction (WLP) algorithm [17] are also presented to locate the NF source.…”

Section: Introductionmentioning

confidence: 99%

“…However, when a source is in the Fresnel region of the array aperture [2][3][4][5], those algorithms in [2][3][4][5][6][7][8][9][10][11] will fail to locate sources. Therefore, for the NF source localization issue, some other algorithms [12][13][14][15][16][17][18][19][20][21][22][23][24][25] are also developed. More specifically, for 2D NF sources localization, Swindlehurst and Kailath proposed a spatial Wigner distribution approach [12], however, this method has a less-than-ideal estimation accuracy.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Localization of 3d Near-Field Source Using the Aperture Extension Method and Nonuniform Cross Array

Jiang¹,

Duan²,

Li³

et al. 2013

PIER B

View full text Add to dashboard Cite

Abstract-Depending on the aperture extension (AE), a high performance three-dimensional (3D) near-field (NF) source localization algorithm is proposed with the nonuniform linear array (NLA). The proposed algorithm first generates some fictitious sensors to extend the array aperture by constructing a new Toeplitz matrix, and then obtains a two-dimensional (2D) covariance matrix which only contains the elevation angle and range parameters, and another 3D covariance matrix which contains the elevation/azimuth angle and range parameters. Then based on the 2D covariance matrix, both the elevation angle and range parameters are estimated by using the NLA along the Z axis. With the estimates of both the elevation angle and range parameters and combining the 3D covariance matrix, the estimates of the azimuth angle parameters are obtained using the NLA along the Y axis. The proposed algorithm has four main merits: i) unlike some classical NF source localization algorithms, the quarter-wavelength sensor spacing constraint is not required and more sources can be located simultaneously by the proposed algorithm; ii) the 3D parameters of the proposed algorithm are paired automatically; iii) the 3D search required in conventional 3D multiple signal classification (MUSIC) algorithm is replaced with only one-dimensional (1D) search, and thus the computational burden is reduced; iv) the proposed algorithm gains superior parameter estimation accuracy and resolution.

show abstract

Computationally Efficient Radio Frequency Source Localization for Radio Interferometric Arrays

2018

View full text Add to dashboard Cite

Radio frequency interference (RFI) is an ever‐increasing problem for remote sensing and radio astronomy, with radio telescope arrays especially vulnerable to RFI. Localizing the RFI source is the first step to dealing with the culprit system. In this paper, a new localization algorithm for interferometric arrays with low array beam sidelobes is presented. The algorithm has been adapted to work both in the near field and far field (only the direction of arrival can be recovered when the source is in the far field). In the near field the computational complexity of the algorithm is linear with search grid size compared to cubic scaling of the state‐of‐the‐art 3‐D MUltiple SIgnal Classification (MUSIC) method. The new method is as accurate as 3‐D MUSIC. The trade‐off is that the proposed algorithm requires a once‐off a priori calculation and storing of weighting matrices. The accuracy of the algorithm is validated using data generated by low‐frequency array while a hexacopter was flying around it and broadcasting a continuous‐wave signal. For the flight, the mean distance between the differential GPS positions and the corresponding estimated positions of the hexacopter is 2 m at a wavelength of 6.7 m.

show abstract

Near-field multiple source localization by passive sensor array

Cited by 355 publications

References 14 publications

Near-field Source Localization Method using Matrix Pencil

Near-field Source Localization Method using Matrix Pencil

Localization of 3d Near-Field Source Using the Aperture Extension Method and Nonuniform Cross Array

Computationally Efficient Radio Frequency Source Localization for Radio Interferometric Arrays

Contact Info

Product

Resources

About